Decision Makers: Avoiding the Pitfalls of Cost Estimate Classification
Category: Thought Leadership
Late last year, Ontario’s provincial government cancelled a project to bring Light Rail Transit (LRT) to the City of Hamilton. This angered residents and local politicians alike who had long been promised better transportation options in the region and the attractive socioeconomic benefits that come with them.
Planning for the project has been in the works for years. Project organizations were stood up, designs progressed and billions of dollars committed. Properties had been purchased to make room for the line along the proposed corridor, and investors jumped at real estate opportunities in areas of town that would modernize. It was a bad day all around, with eroded trust and reputational damage realized by many decision makers in the Greater Toronto and Hamilton Area transit arena.
The reason given for the cancellation? A growing price tag. I would propose that the price tag probably didn’t “grow,” per se (as my colleague @carrieokizaki likes to say: Eventually projects just cost what they cost). But there is little doubt in my mind the cost estimate and the risk profile were certainly mischaracterized and misunderstood at the highest levels of the project sanctioning authority right up to the point the ejector seat was pulled.
I see elements of this in every industry that executes big projects, and while decision makers will always be influenced by optimism bias, it is the role of the project estimating and risk teams to push back with clear and solid basis when they see it happening. We have tools such as estimate classification and cost and schedule contingency confidence levels (PXX) to help us achieve that. With these tools, there should be few surprises – none to the degree of that we observed for the Hamilton LRT.
Understanding Estimate Classification
Project Manager: “345 million. It’s a Class 3 estimate with a P90 contingency.”
Decision Maker: “Well, that sounds pretty good. 345 million it is!”
Estimate classification is a valuable tool, introduced to help project teams and decision makers understand the relationship between the level of project definition and maturity and the degree of certainty attributed to a project cost estimate. If used correctly, it is a powerful framework to contextualize project cost expectations.
However, in some organizations, the principles of cost estimating can be loosely applied and poorly understood. The result is that the defined estimate class becomes a quick and digestible substitute for a more thorough and comprehensive understanding of the actual Basis of Estimate. Similar to the way Monte Carlo confidence levels can be misinterpreted when estimating contingency and management reserve needs, estimate classification can become a prepackaged term that could misrepresent the actual detailed underpinnings of the cost estimate.
As a result, executives may make go or no-go decisions with a false sense of security, without full transparency or clarity into what they’re approving. As another colleague @ericgould likes to say: There is no referee from the Association for Advancement of Cost Engineering that is going to throw a penalty flag on the field if someone isn’t using an appropriate estimate class.
So, it really is up to the decision makers sanctioning the projects to get behind the curtain and ensure they know what went into the Basis of Estimate and confirm that it makes sense. They need to trust but verify, and interrogate things that don’t make sense.
Estimate classification is an excellent and powerful tool that isn’t going anywhere, and rightfully so. But there is no substitute for getting down into the Basis of Estimate and making sure that it was built with the best available information – lessons learned, real data from similar projects and other factors project managers would use to judge the reliability of the estimate if there was no classification system to use.
Think of it this way: If you didn’t have the estimate class framework to rely on, and you just wanted to know how solid this estimate really is, what are the factors you would evaluate? Does the estimate make sense? What comparables did you use? Did you incorporate data from recent similar projects? How advanced is the design? Did you do the geotechnical assessments? Who was involved? Did you get external expertise and input? Is there triangulation of scope, schedule, and cost? What assumptions did you make? Are they reasonable? What are the risks? How do you know? What will the impact of estimate uncertainty be on the bottom line?
These are a few of the things that are important in making solid project decisions, no matter how the estimate is labeled.